Effects on wrong-patient errors by limiting access to concurrently open ERH charts: A preliminary systematic mapping and synthesis review

BackgroundSeveral recent outcome studies have been published looking at the effects of restricting electronic health record (EHR) user interfaces to limit the number of concurrently accessible patient records. Strong recommendations have been in place for several years to have user interfaces constrained to only display one patient chart at a time in order to reduce the risk of data (documentation, orders) being entered on the wrong patient (Joint Commission, 2015; ONC, 2016). This recommendation was made based on expert opinion rather than objective information, raising the question whether the accumulating evidence supports continued implementation of such chart access restrictions. ObjectivesThis work reports a systematic mapping and synthesis review addressing research questions, "What is the evidence that restricting the number of concurrently open records reduces errors? (RQ1), "How effective is restriction of concurrently open charts at reducing wrong-patient errors? (RQ2), and "What additional inquiry is needed to make evidence-based policy decisions about restricting concurrent chart access? (RQ3). MethodsA systematic search of CINAHL, PubMed, and Web of Science databases was performed with full search string specification to retrieve a result set that is the conjunction of result sets for concepts of EHR, concurrently open charts, and medical error. Of 407 studies identified and screened, five were eligible for inclusion in the qualitative synthesis review, and three were amenable to data extraction and pooled effect size calculation. ResultsNone of the studies included for review found evidence of statistically significant change in wrong-patient error rates associated with implementing restriction in the number of patient records allowed to be open concurrently in the EHR. The combined OR for the pooled studies was 1.02 (95% CI 0.90 - 1.15) with low estimates for inter-study heterogeneity and no indication of publication bias. ConclusionThere is no evidence that restricting the number of concurrently open records reduces errors (RQ1). It is not possible to definitively answer RQ2, but the magnitude of any yet to be detected beneficial effect that might be lost with lifting of chart access restriction can be no greater than an absolute risk increase of 33 errors per 100,000 ordering sessions. While it has been claimed that restricting the number of concurrently open EHR records is necessary for patient safety, the present review demonstrates that it is insufficient to attain a measurable improvement in error rates. Additional investigation of the usability and human factors aspects of EHR configuration decisions as well as knowledge of the impacts on clinical workflows will be necessary to provide policymakers, operational leaders, and practitioners with insight into the nature of the threats and opportunities with respect to safety, as well as the strengths and weaknesses of potential interventions.